Complex diseases are controlled by a large number of potentially interacting genetic and environmental factors. Each year millions of Americans are diagnosed with complex diseases such as diabetes, heart disease, Alzheimer's, and various forms of cancer. If the precise genetic loci contributing to variation in disease risk could be characterized, disease incidence might be predicted, and treatments could be modified on a case-by-case basis to halt disease progression and ameliorate suffering (so-called personalized medicine). Unfortunately, progress towards identifying the catalog of genetic risk alleles has been slow, in part because it is unclear what we are searching for: Risk alleles may be common in the population, individually conferring only slight disease risk, or may be rare, but with much larger effects. Our proposal explores a novel approach to characterize genetic loci that combines the strengths of typical QTL (Quantitative Trait Locus) and association mapping, while minimizing some of their weaknesses. A feature of the approach is that it provides a small set of candidate causative sites for each QTL mapped: Previous methodologies have had great difficulty moving from QTL to the precise nucleotides involved. The main aim of the proposal is to develop an integrated framework for the genetic dissection of complex traits in the model organism Drosophila melanogaster. A pair of synthetic laboratory populations will each be initiated with eight inbred founder lines, and following many generations of maintenance a large panel of RILs (Recombinant Inbred Lines) will be derived. The genome of each RIL will be a fine-scale mosaic of segments inherited from the eight founder strains. Following genotyping the panel will serve as a valuable community resource in which to map QTL with high resolution, and we will facilitate these user-directed experiments by creating a web-based analysis portal. The approach yields considerable power to map QTL relative to commonly applied association study methods, and offers the extremely important ability to estimate the frequency of mapped QTL, and in turn the contribution of common alleles to complex trait variation. By resequencing the genomes of the founder strains, for each QTL one can additionally identify the small handful of putatively causative DNA variants to target for future work. We will map QTL for four stress- and drug-tolerance demonstration traits, chosen for their ease of measurement and modest heritability. Each trait will be scored in three mapping designs: directly in homozygous RILs, in the heterozygous progeny of round-robin RIL-RIL crosses, and after crossing RILs to a common inbred reference strain. The two heterozygous testcrosses will examine the effect of inbreeding on QTL mapping directly in RILs. These testcrosses are relatively simple to carry out in flies, and the results may have important implications for the design of experiments using the mouse "Collaborative Cross" eight-way RILs, a system in which the testcrosses would be considerably more cumbersome and expensive to perform.